Super-resolution Localization Microscopy With Large Field Of View

Recently, super-resolution localization microscopy (SRLM) achieves up to20nm spatial resolution, and thus becomes one of the unprecedented tools for cell biology, microbiology and neurobiology. SRLM would be of great beneficial for the study of cell communication and cell signal transduction, especially in visualizing the connectivity of neural circuits, if we can maintain the ultrahigh spatial resolution of the technique, while significantly increase its image field of view (FOV). Unfortunately, the FOV of current SRLM is still very limited. This study focuses on the relationship among spatial resolution, temporal resolution and FOV in SRLM, and is divided into four parts described as following.(1) The key factors determining the spatial resolution of SRLM. We analyzed the theoretical models for signal photon noise, pixelation noise, and background noise. With simulation and experiments, we evaluated the accuracy and applicability of current models for calculating localization precision. We found that the background noise in biology sample is proportional to the second power of pixel size when low-light detector is used, which is not correctly included in current models for calculating localization precision. We further found that the optimal pixel size in SRLM is80nm when it is required to obtain a localization precision of20nm localization in certain signal (850photons) and background (3150photon/μm2) levels.(2) The key factors determining the FOV of SRLM. We analyzed how to improve FOV while maintain a certain temporal resolution. We found that:(a) To obtain large FOV illumination, total internal reflection fluorescence (TIRF), highly inclined and laminated optical sheet (HILO), and high power laser are all required; and (b) To obtain large FOV detection, high collection efficiency optics, and high-performance low-light detectors with fast frame rates and large pixel array size.(3) Super-resolution localization microscopy with sCMOS camera. We characterized the contribution of different noise sources to the total noise in single molecule imaging, and found out that SRLM is working in the shot noise limited region with a combination use of sCMOS camera and bright fluorescence probe (such as d2EosFP). We further demonstrated experimentally that sCMOS camera is capable of imaging actin bundles, evidencing the capability of utilizing sCMOS camera in SRLM.(4) A pilot study of super-resolution localization microscopy with large field of view. Basing on the parallel-readout nature of sCMOS camera, we built a SRLM system with FOV as large as210x185μm2, which is-40times larger than the FOV in current SRLM. Furthermore, we investigated the relation between spatial resolution and the total number of image frames, and found out that such system with large FOV can be realized with spatial resolution of70nm and temporal resolution of5s.